General Idea

Hypotheses

Main conclusions:

  • The bivariate correlation of symbolic ideology with affective ambivalence is negative, which supports the Jost-perspective.
  • The bivariate correlation of political interest with affective ambivalence is negative, which supports the idea that elaborating on a topic tends to deacrease ambivalence.
  • Measures of psychological needs (open_cons as a measure of needs for security and certainty) and cognitive style (nfc_M) are very weakly correlated with affective ambivalence and there is also no evidence for a u-shaped association.
  • The measure of needs for security and certainty (open_cons) is more strongly correlated with symbolic ideology than the measure of need for cognitive closure (nfc_M).
  • Political interest drops at the middle category of symbolic ideology creating a u-shaped pattern.
  • The inversely u-shaped association of symbolic ideology with affective ambivalence holds when controlling for political interest.
  • The inversely u-shaped association of ideology with affective ambivalence can be observed for economic and social ideology.

Ideas for further analyses:

  • Test robustness of the ideology-ambivalence link in multivariate analyses with controls.

Explore correlations

## 
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
Wave 4
term symb_id_w4 open_cons nfc_M pol_int_w4 Merkel_amb_w4 Schulz_amb_w4 mean_amb_w4
symb_id_w4 NA 0.1465046 0.0786502 -0.0740616 -0.0665105 -0.0796847 -0.0826058
open_cons 0.1465046 NA 0.3832139 -0.1068767 -0.0244760 -0.0097061 -0.0199300
nfc_M 0.0786502 0.3832139 NA -0.0670255 -0.0282233 -0.0122005 -0.0227356
pol_int_w4 -0.0740616 -0.1068767 -0.0670255 NA -0.0797389 -0.1138886 -0.1120108
Merkel_amb_w4 -0.0665105 -0.0244760 -0.0282233 -0.0797389 NA 0.4683591 0.8722972
Schulz_amb_w4 -0.0796847 -0.0097061 -0.0122005 -0.1138886 0.4683591 NA 0.8460049
mean_amb_w4 -0.0826058 -0.0199300 -0.0227356 -0.1120108 0.8722972 0.8460049 NA 
## 
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
Wave 6
term symb_id_w6 open_cons nfc_M pol_int_w6 Merkel_amb_w6 Schulz_amb_w6 mean_amb_w6
symb_id_w6 NA 0.1512384 0.0794083 -0.0537975 -0.1001022 -0.0863655 -0.1085498
open_cons 0.1512384 NA 0.3832139 -0.0996671 -0.0259161 -0.0102304 -0.0216856
nfc_M 0.0794083 0.3832139 NA -0.0693905 -0.0309360 -0.0100961 -0.0241923
pol_int_w6 -0.0537975 -0.0996671 -0.0693905 NA -0.0790483 -0.1021481 -0.1060417
Merkel_amb_w6 -0.1001022 -0.0259161 -0.0309360 -0.0790483 NA 0.4574905 0.8669063
Schulz_amb_w6 -0.0863655 -0.0102304 -0.0100961 -0.1021481 0.4574905 NA 0.8423697
mean_amb_w6 -0.1085498 -0.0216856 -0.0241923 -0.1060417 0.8669063 0.8423697 NA

U-shape: Symbolic ideology & affective ambivalence

Analytic approach:

Inversely u-shaped associations of variables with affective ambivalence could be tested using the following approaches:

  • Two lines test
  • Polynomial regression
  • Lonlinear moderation of the association of positive and negative affect

Wave 4

Average ambivalence

Two lines test

Polynomial regression

Model 1Model 2Model 3Model 4
(Intercept)0.37 ***(0.00)0.30 ***(0.01)0.44 ***(0.01)0.38 ***(0.01)
rescale(symb_id_w4)-0.08 ***(0.01)0.29 ***(0.03)-0.09 ***(0.01)0.25 ***(0.03)
I(rescale(symb_id_w4)^2)           -0.40 ***(0.03)           -0.36 ***(0.03)
gendfemale                      0.01    (0.00)0.00    (0.00)
age                      -0.00 ***(0.00)-0.00 ***(0.00)
edumedium                      0.01    (0.01)0.01    (0.01)
eduhigh                      0.00    (0.01)0.00    (0.01)
inc                      0.00    (0.00)-0.00    (0.00)
regionWest Germany                      0.01    (0.01)0.01    (0.01)
N10081           10081           9777           9777           
R20.01        0.02        0.02        0.03        
*** p < 0.001; ** p < 0.01; * p < 0.05.

Merkel

Two lines test

Polynomial regression

Model 1Model 2Model 3Model 4
(Intercept)0.36 ***(0.01)0.29 ***(0.01)0.42 ***(0.01)0.35 ***(0.02)
rescale(symb_id_w4)-0.08 ***(0.01)0.30 ***(0.04)-0.08 ***(0.01)0.26 ***(0.04)
I(rescale(symb_id_w4)^2)           -0.41 ***(0.04)           -0.37 ***(0.04)
gendfemale                      0.00    (0.01)0.00    (0.01)
age                      -0.00 ***(0.00)-0.00 ***(0.00)
edumedium                      0.02 ** (0.01)0.02 ** (0.01)
eduhigh                      0.02 *  (0.01)0.02 *  (0.01)
inc                      0.00    (0.00)0.00    (0.00)
regionWest Germany                      0.01    (0.01)0.00    (0.01)
N10048           10048           9745           9745           
R20.00        0.02        0.02        0.02        
*** p < 0.001; ** p < 0.01; * p < 0.05.

Schulz

Two lines test

Polynomial regression

Model 1Model 2Model 3Model 4
(Intercept)0.38 ***(0.01)0.31 ***(0.01)0.46 ***(0.01)0.40 ***(0.02)
rescale(symb_id_w4)-0.09 ***(0.01)0.28 ***(0.04)-0.09 ***(0.01)0.24 ***(0.04)
I(rescale(symb_id_w4)^2)           -0.40 ***(0.04)           -0.36 ***(0.04)
gendfemale                      0.01 *  (0.00)0.01    (0.00)
age                      -0.00 ***(0.00)-0.00 ***(0.00)
edumedium                      -0.00    (0.01)-0.00    (0.01)
eduhigh                      -0.01    (0.01)-0.01    (0.01)
inc                      -0.00    (0.00)-0.00    (0.00)
regionWest Germany                      0.01 *  (0.01)0.01 *  (0.01)
N9833           9833           9535           9535           
R20.01        0.02        0.02        0.03        
*** p < 0.001; ** p < 0.01; * p < 0.05.

Wave 6

Average ambivalence

Two lines test

Polynomial regression

Model 1Model 2Model 3Model 4
(Intercept)0.37 ***(0.00)0.31 ***(0.01)0.44 ***(0.01)0.38 ***(0.01)
rescale(symb_id_w6)-0.10 ***(0.01)0.26 ***(0.03)-0.11 ***(0.01)0.22 ***(0.03)
I(rescale(symb_id_w6)^2)           -0.39 ***(0.03)           -0.35 ***(0.03)
gendfemale                      0.00    (0.00)-0.00    (0.00)
age                      -0.00 ***(0.00)-0.00 ***(0.00)
edumedium                      0.01 ** (0.01)0.01 ** (0.01)
eduhigh                      0.01    (0.01)0.01    (0.01)
inc                      0.00    (0.00)0.00    (0.00)
regionWest Germany                      0.01    (0.00)0.01    (0.00)
N12005           12005           11710           11710           
R20.01        0.03        0.03        0.04        
*** p < 0.001; ** p < 0.01; * p < 0.05.

Merkel

Two lines test

Polynomial regression

Model 1Model 2Model 3Model 4
(Intercept)0.37 ***(0.01)0.30 ***(0.01)0.43 ***(0.01)0.36 ***(0.01)
rescale(symb_id_w6)-0.12 ***(0.01)0.30 ***(0.03)-0.12 ***(0.01)0.26 ***(0.03)
I(rescale(symb_id_w6)^2)           -0.45 ***(0.04)           -0.41 ***(0.04)
gendfemale                      -0.00    (0.00)-0.01    (0.00)
age                      -0.00 ***(0.00)-0.00 ***(0.00)
edumedium                      0.03 ***(0.01)0.02 ***(0.01)
eduhigh                      0.02 ***(0.01)0.02 ***(0.01)
inc                      0.00 *  (0.00)0.00    (0.00)
regionWest Germany                      0.00    (0.01)0.00    (0.01)
N11976           11976           11681           11681           
R20.01        0.02        0.02        0.04        
*** p < 0.001; ** p < 0.01; * p < 0.05.

Schulz

Two lines test

Polynomial regression

Model 1Model 2Model 3Model 4
(Intercept)0.38 ***(0.01)0.32 ***(0.01)0.45 ***(0.01)0.40 ***(0.01)
rescale(symb_id_w6)-0.09 ***(0.01)0.21 ***(0.03)-0.10 ***(0.01)0.17 ***(0.03)
I(rescale(symb_id_w6)^2)           -0.33 ***(0.03)           -0.29 ***(0.03)
gendfemale                      0.00    (0.00)0.00    (0.00)
age                      -0.00 ***(0.00)-0.00 ***(0.00)
edumedium                      0.00    (0.01)0.00    (0.01)
eduhigh                      -0.00    (0.01)-0.00    (0.01)
inc                      -0.00    (0.00)-0.00    (0.00)
regionWest Germany                      0.01 ** (0.01)0.01 *  (0.01)
N11900           11900           11605           11605           
R20.01        0.02        0.02        0.03        
*** p < 0.001; ** p < 0.01; * p < 0.05.

U-shape: Symbolic ideology & Need for Closure

Wave 4

Two lines test

Wave 6

Two lines test

U-shape: Symbolic ideology & Needs for Security and Certainty

Wave 4

Two lines test

Wave 6

Two lines test

U-shape: Symbolic ideology & Political Interest

Wave 4

Two lines test

Wave 6

Two lines test

U-shape: Affective Ambivalence & Political Interest

Wave 4

Average ambivalence

Two lines test

Merkel

Schulz